Viewing multi-dimensional data through hierarchical visualization

• US 20030200191A1
• Filed: 03/28/2003
• Published: 10/23/2003
• Est. Priority Date: 04/19/2002
• Status: Active Grant

• Current Assignee: Computer Associates Think Inc. (Broadcom, Inc.)
• Original Assignee: Computer Associates Think Inc. (Broadcom, Inc.)
• Inventors: Duan, Baofu, Pao, Yoh-Han, Meng, Zhuo
• Granted Patent: US 7,777,743 B2
• Time in Patent Office: Days
• US Class Current: 706/47
• CPC Class Codes: G06F 18/213 Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
• CPC Class Codes: G06F 18/40 Software arrangements specially adapted for pattern recognition, e.g. user interfaces or toolboxes therefor

Abstract

A method for hierarchical visualization of multi-dimensional data is provided. A first dimension-reduction process is applied to a multi-dimensional data set to obtain a first visualization. A subset of the multi-dimensional data set associated with a selected region of the dimension-reduced first visualization is selected. A second dimension-reduction process is applied to the selected subset of the multi-dimensional data set to obtain at least one additional visualization.

Claims

1. A method for hierarchical visualization of multi-dimensional data, comprising:

(a) applying a first dimension-reduction process to a multi-dimensional data set to obtain a first visualization;

(b) selecting a subset of the multi-dimensional data set associated with a selected region of the dimension-reduced first visualization where more detail is desired; and

(c) applying a second dimension-reduction process to the selected subset of the multi-dimensional data set to obtain at least one additional visualization.

2. The method of claim 1, wherein (b) and (c) are repeated for a further subset associated with a subregion of the selected region.

3. The method of claim 1, wherein the subregion includes a mixed portion.

4. The method of claim 1, wherein (b) and (c) are repeated, until a dimension-reduced visualization having a desired level of separation of points is obtained.

5. The method of claim 4, wherein each visualization has an associated angle of view and a view of the multi-dimensional data set is obtained at an angle of view associated with the dimension-reduced visualization having the desired level of separation of points.

6. The method of claim 1, wherein each visualization has an associated angle of view and a view of the multi-dimensional data set is obtained at an angle of view associated with a selected one of the at least one additional visualization.

7. The method of claim 1, wherein the first visualization has a first angle of view, and the additional visualization has a second angle of view different from the first angle of view.

8. The method of claim 1, wherein the at least one additional visualization is at a higher level of detail than the first visualization.

9. The method of claim 1, wherein the second dimension-reduction process applies the same dimension-reduction technique as used in the first dimension-reduction process.

10. The method of claim 1, wherein the first dimension-reduction process and the second dimension-reduction process apply respectively different dimension-reduction techniques.

11. The method of claim 1, wherein the second dimension-reduction process includes applying a continuous dimension-reduction technique to obtain a sequence of dimension-reduced visualizations.

12. The method of claim 11 further comprising

selecting two data points in the multi-dimensional data set for distance estimation,

wherein if the two data points appear to be far apart in any one of the dimension-reduced visualizations, the two data points are far apart in the original multi-dimensional space.

13. The method of claim 12, wherein the continuous dimension-reduction technique includes principal component analysis, and a largest one of the reduced-dimension distances is a lower bound estimate of an actual distance between the two selected data points in the original multi-dimensional space.

14. The method of claim 1 further comprising utilizing a hierarchical cluster tree to automate generation of hierarchical visualizations by generating a visualization for each node of the cluster tree.

15. The method of claim 1, wherein the subset selected in (b) corresponds to a mixed region.

16. The method of claim 1, wherein the multi-dimensional data set includes non-numeric data and is preprocessed into numerical form prior to dimension reduction.

17. The method of claim 1, wherein the additional visualization is consulted with data from a test set by applying a mapping corresponding to the second dimension-reduction process.

18. The method of claim 1, wherein the method is applied to classify the multi-dimensional data set according to one or more features associated with the data set.

19. The method of claim 1, wherein the multi-dimensional data set is collected from a production process, and the method is applied to obtain information for predicting product properties.

20. The method of claim 1, wherein the multi-dimensional data set corresponds to data collected from a system, and the method is applied to obtain information for diagnosing a problem in the system.

21. The method of claim 1, wherein the multi-dimensional data set corresponds to data collected from a system, and the method is applied to obtain information for predicting a problem, before the problem develops in the system.

22. The method of claim 1, wherein the multi-dimensional data set corresponds to data collected from a system, and the method is applied to obtain information for optimizing the system.

23. The method of claim 1, wherein the multi-dimensional data set corresponds to data collected from a system, and the method is applied to obtain information for searching the system.

24. A program storage device readable by a machine, tangibly embodying a program of instructions executable by the machine to perform steps in a method for hierarchical visualization of multi-dimensional data, the method steps comprising:

(a) applying a first dimension-reduction process to a multi-dimensional data set to obtain a first visualization;

(b) selecting a subset of the multi-dimensional data set associated with a selected region of the dimension-reduced first visualization; and

(c) applying a second dimension-reduction process to the selected subset of the multi-dimensional data set to obtain at least one additional visualization.

25. The program storage device of claim 24, wherein (b) and (c) are repeated, until a sufficiently detailed visualization having a desired level of separation of points is obtained.

26. The program storage device of claim 24, wherein (b) and (c) are repeated for a further subset associated with a subregion of the selected region.

27. A computer system, comprising:

a processor; and

a program storage device readable by the computer system, tangibly embodying a program of instructions executable by the processor to perform steps in a method for hierarchical visualization of multi-dimensional data, the method steps comprising;

applying a first dimension-reduction process to a multi-dimensional data set to obtain a first visualization;

selecting a subset of the multi-dimensional data set associated with a selected region of the dimension-reduced first visualization; and

applying a second dimension-reduction process to the selected subset of the multi-dimensional data set to obtain at least one additional visualization.

28. The computer system of claim 27, wherein (b) and (c) are repeated, until a sufficiently detailed visualization having a desired level of separation of points is obtained.

29. The computer system of claim 27, wherein (b) and (c) are repeated for a further subset associated with a subregion of the selected region.

30. A computer data signal embodied in a transmission medium comprising:

a first segment including dimension-reduction code to perform a dimension-reduction process to a multi-dimensional data set to obtain a first visualization; and

a second segment including data selection code to select a subset of the multi-dimensional data set associated with a selected region of the dimension-reduced first visualization

wherein the dimension-reduction code is applied at least one time to the selected subset of the multi-dimensional data set to obtain at least one additional visualization.